Physical therapy no-show rates: benchmarks and fixes

Somebody finally assembled the dataset PT owners always say they want: 444,995 patients, more than six million scheduled visits, 697 outpatient clinics across 26 states. The result, published in PLOS One, is blunt. 73% of patients missed at least one appointment during their episode of care. Nearly three of every four plans of care leaked a visit somewhere.

One definition note before you quote that number at a staff meeting. The study counted a visit as missed when it was scheduled but ended canceled or pending, so late cancellations are in there too. Keep that in mind, because it is the honest version of the shrug on every software blog: “no-show rates range from 10% to over 70%.” The range is useless. The mechanics underneath it are not.

What is a normal no-show rate for physical therapy?

Across outpatient care, per-visit miss rates mostly land between 10% and 20%, and in physical therapy specifically, 73% of patients miss at least one appointment over a full episode of care. The per-visit figure comes from a BMC Health Services Research study of VA outpatient clinics. Over a decade of appointments, the mean no-show rate was 18.8% across ten specialties. Under 10% at a private clinic is a tight ship.

The per-episode framing is the one that should bother you, because PT is repeat-visit care. A primary care office loses one slot when a patient skips. Your clinic loses the slot plus a piece of a 12-visit plan of care, and the same study found each miss makes the next one likelier. Attendance is a clinical outcome wearing a scheduling costume.

Your schedule already knows who will miss next

In the PLOS One data, the strongest predictor of a missed visit was simple. It was the number of appointments the patient had already canceled. After that: how far in advance the visit was booked, and how long it had been since the last one. Medicaid coverage and comorbidity count showed up too, but the top three are the ones your front desk controls.

That list converts directly into policy. Book the next visit before the patient leaves the building, since lead time you never create cannot rot. Treat a second cancellation as an event: the front desk calls, asks what changed, and rebooks to a time that actually fits. And when a gap opens mid-plan, close it that week; a patient who has been away three weeks is drifting toward the door. None of this requires software, though a schedule that surfaces those flags means nobody has to remember to go looking for them.

Do appointment reminders actually work?

Yes, and the measured effect is bigger than most owners assume. A BMJ Open meta-analysis of 21 randomized studies found patients who received text-message reminders were 23% more likely to attend. They were also 25% less likely to no-show than patients who received no reminder at all. The detail practices skip past: multiple reminders beat one. In the pooled data, attendance reached 78% with a reminder sequence against 62% with a single message.

So if your system fires one text the night before, most of the measured effect is still on the table. Run a sequence: a confirmation at booking, a reminder a few days out, another the day before.

In Orion that sequence runs itself off the schedule and a reply lands back on the front desk queue, but the evidence holds whichever system sends the message. What does not hold up is making the front desk the reminder engine. The days they are slammed are exactly the days the calls never happen.

Should you charge a no-show fee?

A PT practice can charge a no-show fee, and that includes charging Medicare patients. CMS has allowed missed-appointment charges since October 2007 under a policy spelled out in MLN Matters article MM5613. CMS treats the fee as “a charge for a missed business opportunity,” not a medical service, and permits it on two conditions. Your policy must charge Medicare and non-Medicare patients the same amount, and the fee goes to the patient directly, never on a claim to Medicare.

Whether you should lean on it is a different question. A fee triggers after the miss, and everything predictive in the attendance data fires before it: the long lead time, the second cancellation, the widening gap. Publish the policy in writing at the first visit so it is credible, then spend the real effort upstream.

The $200 number, and the one that matters

About the “$200 per no-show” figure that circulates on scheduling software blogs. The closest thing to a primary source behind it is that same VA study: an average cost of $196 per missed appointment, in 2008 dollars, at Veterans Affairs medical centers. Whatever your misses cost, it is not a number measured at a VA hospital eighteen years ago.

Your version takes two minutes. Average collections per visit, times the misses you failed to backfill last week, times 52. A clinic collecting $95 a visit and eating ten unfilled slots a week is losing roughly $49,000 a year, and unlike a denied claim there is no appeal. We made the same argument about denial benchmarks: the published figures are context, and the number that changes behavior comes out of your own reports. Run yours, tape it to the front desk monitor, and the reminder sequence stops being an IT task. It becomes the cheapest hire you never made.